Photoinduced intramolecular electron transfer in a 2,7-diaminofluorene chromophore decorated with two benzophenone subunits

Article abstract of DOI:10.1039/b813717j

An extensive photophysical analysis of a 2,7-bis-(N-4-methoxyphenyl-N- phenylamino)fluorene derivative covalently linked with two benzophenone moieties is presented. A systematic comparison with a model chromophore without benzophenone was performed. For

Full text of DOI:10.1039/b813717j

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/pccp | Physical Chemistry Chemical Physics
Photoinduced intramolecular electron transfer in a 2,7-diaminofluorene
chromophore decorated with two benzophenone subunits
a
a
a
b
´
Ming Jin, Jean-Pierre Malval,* Fabrice Morlet-Savary, Helene Chaumeil,
Albert Defoin,b Pinar Batatc and Gediminas Jonusauskasc
Received 7th August 2008, Accepted 12th January 2009
First published as an Advance Article on the web 16th February 2009
DOI: 10.1039/b813717j
An extensive photophysical analysis of a 2,7-bis-(N-4-methoxyphenyl-N-phenylamino)fluorene
derivative covalently linked with two benzophenone moieties is presented. A systematic
comparison with a model chromophore without benzophenone was performed. For both
chromophores, the electronic properties of the ground states are completely equivalent indicating
that benzophenone subunits do not exhibit any electronic interaction with the diaminofluorene
core. However, at the singlet excited state, the presence of benzophenones induces the occurrence
of additional non-radiative de-excitation pathways. Even the intersystem crossing rate is
significantly increased with respect to that of the model one. A photoinduced intramolecular
electron transfer (PIET) from diaminofluorene to benzophenone subunits is proposed as the most
efficient quenching process. At low polar solvent, the emission of an exciplex confirms the PIET
process and the occurrence of a partial charge separation between donor and acceptor parts.
such a charge recombination is the use of donor and acceptor
Introduction
subunits that can generate intrinsic long-lived radical ions.
Focusing on the donor part, the 2,7-diaminofluorene system
Photoinduced intramolecular electron transfer (PIET) leading
to a charge separation in a supramolecular structure plays a
fundamental role in the dynamics of concomitant processes
involved in the photosynthesis, in biology, in photonics or in the
solar energy conversion. The simple association of donor and
acceptor subunits constitutes an elementary model to address
factors inherent to the PIET mechanism.1–6 In this case, the
electronic structure and the geometric conformation of the
excited state are the most important issues. Among the multiple
appears as
a potential candidate. This ‘push–pull–push’
chromophore has attracted attention because of its high interest
in nonlinear optical applications.15–19 Moreover, it is an
electron-rich organic compound that should produce
resonance-stabilized radical cation upon oxidation reaction.
a
Hence in this paper, we characterize the photophysical
properties of a 2,7-bis-(N-4-methoxyphenyl-N-phenylamino)-
fluorene derivative linked with two 3-methoxybenzophenone
subunits as electron acceptor groups (Scheme 1). The nature
and the symmetry of electronic transitions at the ground and
excited states are identified from a detailed spectroscopic
comparison with the photophysical feature of a model
diaminofluorene chromophore without benzophenone moiety.
The electronic and conformational influences of the acceptor
parameters that tune the efficiency of PIET, the distance (dAB
)
between the donor (D) and the acceptor (A) parts constitutes
one of the main decisive variable.7 Its influence has been
extensively highlighted.8–11 For instance, the intramolecular
electron transfer rate constant (keT) in a donor acceptor systems
that are covalently linked with a rigid non-conjugating
spacer12–14 shows an exponential decrease with dAB according
to the empirical relation: keT = Aexp(ꢀb(dAB ꢀ d0AB)). In this
expression, d0AB denotes the distance of closest approach and b
holds information relating to the nature of the linkage and the
medium. Such a formalism is less clearly demonstrated for
donor–acceptor structures linked by flexible spacers since
electron transfer may occur from a wide distribution of excited
conformers.7 Besides, the conformational dynamics at the
excited state as well as molecular environment may promote
the charge recombination. One possible strategy to hamper
a
´
Departement de Photochimie Generale, UMR CNRS 7525,
Universite de Haute Alsace, ENSCMu, 3 rue Alfred Werner,
´ ´
´
68093 Mulhouse Cedex, France. E-mail: jean-pierre.malval@uha.fr
b Laboratoire de Chimie Organique et Bioorganique,
´
UMR CNRS 7015, Universite de Haute Alsace, ENSCMu,
3 rue Alfred Werner, 68093 Mulhouse Cedex, France
c
´
Centre de Physique Moleculaire Optique et Hertzienne,
UMR CNRS 5798, Universite Bordeaux I, 33405 Talence Cedex,
´
France
Scheme 1 Molecular structures of DAFB and DAF.
ꢁc
This journal is the Owner Societies 2009
2622 | Phys. Chem. Chem. Phys., 2009, 11, 2622–2630

View Article Online
groups toward the photophysics of the chromophore are then
discussed.
1595, 1656, 2852, 2924 cmꢀ1
.
1H NMR (C6D6, 400 MHz,
295 K): 0.89 (t, J = 7,0 Hz, 6 H), 0.95 (m, 4 H), 1.11
(m, 8 H), 1.24 (m, 20 H), 1.71 (m, 4 H), 3.76 (m, 8 H), 6.77
(d, J = 8,8 Hz, 4 H), 6.88 (t, J = 7,3 Hz, 2 H), 6.96 (m, 2 H),
6.98 (t, J = 8,0 Hz, 2 H), 7.05 (t, J = 7,4 Hz, 4 H), 7.11–7.18
(m, 12 H), 7.28 (m, 6 H), 7.35 (d, J = 1,5 Hz, 2 H), 7.41
(d, J = 8,1 Hz, 2 H), 7.50 (s, 2 H), 7.78 (d, J = 7,0 Hz, 4 H).
13C NMR (C6D6, 100.6 MHz, 295 K): 14.4, 23.1, 24.6, 29.8,
29.9, 30.1, 30.2, 30.5, 32.3, 40.5, 55.4, 66.67, 66.70, 114.9, 115.8,
118.6, 119.8, 120.3, 122.2, 123.0, 123.3, 127.2, 128.4, 129.5,
129.6, 130.3 (2C), 132.2, 136.3, 138.3, 139.6, 142.0, 147.4, 149.1,
152.5, 155.6, 159.2, 195.5. HRMS, ESI-MicroTOF, calc for
C87H92N2O6, M+: 1260.6955; found: 1260.6994.
Experimental
Materials
The synthetic route for the synthesis of both chromophores is
depicted in Scheme 2. DAF is synthesized from a palladium-
catalysed cross-coupling reaction between 2,7-dibromo-9,9-
didecyl-9H-fluorene and N-(4-methoxyphenyl)aniline in
toluene using a procedure previously described.20,21 The
strategy employed for the preparation of DAFB first consists
in a demethylation of DAF using BBr3 in dichloromethane
at ꢀ78 1C which leads to phenolic alcohols. A nucleophilic
substitution of a bromine atom in 1,2-dibromoethane is
carried out in presence of the latter phenolic product
with potassium carbonate. Finally, a second nucleophilic
substitution of bromine atoms is performed with 3-hydroxy-
benzophenone using a similar method and leads to DAFB.
All the solvents employed for photophysical analysis were
Aldrich or Fluka spectroscopic grade.
General techniques. The absorption measurements were
carried out with a Perkin Elmer Lambda 2 spectrometer.
Steady-state fluorescence and phosphorescence spectra were
collected from a FluoroMax-4 spectrofluorometer. Emission
spectra are spectrally corrected, and fluorescence quantum
yields include the correction due to solvent refractive index
and were determined relative to quinine bisulfate in 0.05 molar
sulfuric acid (F = 0.52).22
9,9-Didecyl-N2,N7-bis-4-methoxyphenyl-N2,N7-diphenyl-9H-
fluoren-2,7-diamine, DAF. IR(KBr): 695, 749, 819, 830, 1039,
1180, 1242, 1269, 1314, 1440, 1466, 1494, 1507, 1594, 2852,
.
2925 cmꢀ1 1H NMR (C6D6, 400 MHz, 295 K): 0.90
(t, J = 7.2 Hz, 6 H), 0.93 (m, 4 H), 1.10 (m, 8 H), 1.24
(m, 20 H), 1.69 (m, 4 H), 3.30 (s, 6 H), 6.74 (d, J = 6,8 Hz,
4 H), 6.86 (t, J = 7,2 Hz, 2 H), 7.13–7.17 (m, 10 H), 7.27
(d, J = 7,8 Hz, 4 H), 7.34 (d, J = 1,6 Hz, 2 H), 7.39
(d, J = 8,3 Hz, 2 H).13C NMR (C6D6, 100.6 MHz, 295 K):
14.4, 23.1, 24.6, 29.8, 29.9, 30.1, 30.2, 30.6, 32.4, 40.5, 55.0,
55.3, 115.2, 118.6, 120.3, 122.0, 123.0, 123.2, 127.3, 129.5,
136.3, 141.6, 147.5, 149.2, 152.5, 156.7. HRMS, ESI-MicroTOF,
calc. for C59H72N2O2, M+: 840.5594; found: 840.5671.
The fluorescence lifetimes were measured using a Nano LED
emitting at 372 nm as an excitation source with a nano led
controller module, Fluorohub from IBH, operating at 1MHz.
The detection was based on an R928P type photomultiplier
from Hamamatsu with high sensitivity photon-counting mode.
The decays were fitted with the iterative reconvolution method
on the basis of the Marquardt–Levenberg algorithm.23 Such a
reconvolution technique allows an overall-time resolution down
to 0.15 ns. The quality of the exponential fits was checked using
the reduced w2 (r1.2).
9,9-Didecyl-N2,N7-bis{4-[(3-benzoylphenoxy )-ethoxy]phenyl}-
N2,N7-diphenyl-9H-fluoren-2,7-diamine DAFB. IR(KBr): 695,
723, 1072, 1236, 1279, 1317, 1439, 1466, 1492, 1506, 1579,
Phosphorescence and steady-state anisotropy measurements
were performed in ethanol at 77 K. The samples are placed in a
5-mm diameter quartz tube inside a Dewar filled with liquid
nitrogen. Two Glan–Thompson polarizers are placed in the
excitation and emission beams. The anisotropy r is determined
IVVꢀgIVH
IHV
IHH
as follows: r ¼
with g ¼
Where I is the
IVVþ2gIVH
fluorescence intensity. The subscripts denote the orientation
(horizontal, H, or vertical, V) of the excitation and emission
polarizers, respectively. g is an instrumental correction factor.
The proper calibration of the set-up was checked using a
recent standard method with rhodamine-101 in glycerol.24
Determination of the intersystem crossing quantum yields
of chromophores (Fisc) were performed by a triplet–triplet
energy transfer method. Experiments were carried out by laser
flash photolysis at nanosecond time scale with an Edinburgh
Analytical Instruments LP900 equipped with a 450-W pulsed
Xe arc lamp, a Czerny–Turner monochromator and a fast
photomultiplier. The samples were irradiated with the third
harmonic (l = 355 nm, B10 ns, 5 mJ per pulse) of a Nd/YAG
Powerlite 9010 from Continuum. The sample concentration
was adjusted to get an absorption of B0.3 at the excitation
wavelength and purged with nitrogen for 15 min prior to
photophysical studies. Triplet–triplet energy transfer was
performed with camphorquinone (CQ) as energy acceptor
Scheme 2 (a) Pd2(dba)3, PtBu, KotBu, toluene, 70 1C, 16h, 85%.
(b) Br3, CH2Cl2, ꢀ78 1C, 2h, rt, 87%. (c) BrCH2CH2Br, K2CO3,
EtOH, 60 1C. (d) K2CO3, acetone.
ꢁc
This journal is the Owner Societies 2009
Phys. Chem. Chem. Phys., 2009, 11, 2622–2630 | 2623

View Article Online
system. The bimolecular charge transfer process between
triplet states of the chromophores and CQ was considered as
negligible since the transient absorption spectrum of the long-
lived radical cations of chromophores located in 450–550 nm
range was not detected. Finally the standard system
benzophenone/camphorquinone (BP/CQ) was used to obtain
Fisc. A detailed experimental methodology is described in
detail in ref. 16
Ultrafast transient absorption was carried out using the
following set-up. A Ti:Sapphire laser system emitting pulses of
0.6 mJ and 30 fs at 800 nm and 1 kHz pulse repetition
rate (Femtopower Compact Pro) with home built optical
parametric generator and frequency mixers followed were
used to excite samples at 370 nm. A white light continuum
(390 nm–1000 nm) pulses generated in a 5 mm methanol cell
was used as a probe. The variable delay time between pump
and probe pulses was obtained using a delay line with 0.1 mm
resolution. The solutions were placed in 1 mm circulating cell
made from fused silica. White light signal and reference
spectra were recorded using a two channel fibre spectrometer
(Avantes Avaspec-2048-2). A home-written acquisition and
experiment control program in LabView made possible to
record transient spectra with an average error less than
10ꢀ4 of optical density (OD) for all wavelengths. The temporal
resolution of our set-up was better than 60 fs. A temporal
chirp of probe pulse was corrected by a computer program
with respect to a Lawrencian fit of a Kerr signal generated in a
0.2 mm glass plate used in a place of sample.
Fig. 1 (A) Absorption spectra of DAF (squares) and DAFB (circles)
in acetonitrile. (B) Excitation anisotropies DAF (squares) and DAFB
(circles) in glassy matrix of ethanol at 77 K (lem = 415 nm).
polarity typically correspond to the S0 - S2 and S0 - S1
electronic transitions centered on the diaminofluorene
moiety.16,26 The existence of such transitions can be revealed
by the steady-state excitation anisotropy spectrum which first
displays a constant value of ca. 0.35 ꢂ 0.01 within the
long wavelength absorption band (Fig. 1B) then decreases
and reaches an other plateau with a mean value of ca.
ꢀ0.02 ꢂ 0.02 in the 280–320 nm region in good accordance
with literature.26 Hence, the absorption transition moment of
the last electronic transition is mainly collinear with the
emitting one, and the electronic symmetry of S0 - S2
transition is clearly distinctive from the last one.
The cyclic voltammetry experiments (using a computer-
controlled Princeton 263A potentiostat with a three-electrode
single-compartment cell; a saturated calomel electrode in
methanol used as a reference was placed in a separate
compartment) were performed at 300 K, in Ar-degassed
These results can be correlated with TDDFT calculations.
In Table 2 the energies of the main electronic transitions, the
corresponding oscillator strengths and the main orientation of
their transition dipole moments are reported. Fig. 2 depicts the
HOMO and the LUMO molecular orbitals involved in last
electronic transitions of DAF. The first singlet excitation
energies at 3.33 eV (373 nm) agree very well with the experi-
mental results. The calculations also indicate that the S0 - S1
transition is a pure HOMO–LUMO transition with a high
oscillator strength (f 4 1) and involves a symmetrical charge
delocalization along the 2,7-bis-(diphenylamino)fluorene bar.
Its corresponding transition dipole moment vector (M01) is
mainly collinear to the long axis of the fluorene structure.27,28
The second absorption band at 305 nm corresponds to two
close electronic transitions located at 4.08 eV (304 nm) and
4.11 eV (302 nm). These transitions involve the same set of
molecular orbitals and exhibit a similar electronic symmetry.
In both cases their transition dipole moment vectors
(M02, M03) have their main components in a plane perpendi-
cular to the long axis of the fluorene moiety. Hence the
transitions ascribable to the second absorption band clearly
exhibit a different electronic symmetry with respect to that of
S0 - S1 one.
acetonitrile with
a constant concentration (0.1 M) of
n-Bu4BF4. Ferrocene was used as an internal reference.
The temperature experiments are performed using
a
cuvette holder provided with
a temperature controller
(TC 102) from Quantum Northwest. The temperature range
is ꢀ10 1C to 105 1C.
The optimization of geometrical structures of DAF and
DAFB were carried out in the frame work of the density
functional theory (DFT) with the B3LYP/6-31G* basis and
using the GAUSSIAN 03 suite.25 To prevent long time
calculation and convergence to secondary minima in the
potential-energy surface, the long decyl branches were
replaced by short ethyl ones. From the time-dependent
calculations at TD/MPW1PW91/6-31G* levels, the
absorption properties of each chromophore were calculated.
Results and discussion
Ground-state properties
Fig. 1A displays the superimposition of the absorption spectra
of both compounds in acetonitrile and Table 1 gathers
spectroscopic data in various solvents. The comparison of
the absorption spectra indicates that the red part of the spectra
is quite similar with two predominant bands located at 300 nm
and 375 nm. These bands which are insensitive to solvent
From Fig. 1, we can also notice that the absorption spec-
trum of DAFB differs from the absorption spectrum of its
analogue one below 340 nm. While the last absorption band of
the 3-methoxybenzophenones located in the 300–350 nm
region is mainly obscured by the absorption of the
ꢁc
This journal is the Owner Societies 2009
2624 | Phys. Chem. Chem. Phys., 2009, 11, 2622–2630

View Article Online
2,7-diaminofluorene moiety, the 1La transition of benzo-
phenone derivatives is clearly observed at 250 nm.29,30 The
extinction coefficient relative to the maximum of last absorp-
tion band has a quite similar value for both compounds
(B30 000 Mꢀ1 cmꢀ1), hence the 3-methoxybenzophenone
does not seem to induce any interactions with the diamino-
fluorene moiety at the ground state.
The redox properties of the compounds were characterized
by cyclic voltammetry. Fig. 3 shows the cyclic voltammograms
for the systems in acetonitrile containing 0.1
M of
(n-Bu)4NPF6, the cathodic part corresponds to the reversible
reduction of 3-methoxybenzophenone moiety, its half-wave
potential has a value of ca. ꢀ1.78 V versus a standard calomel
electrode (SCE) in methanol, perfectly comparable to the
reduction potential of benzophenone measured in the same
condition within experimental error of ꢂ10 mV, it thus
confirms the absence of electronic interaction between the
benzophenone moiety and the donor part. At high potential,
each chromophore shows exactly the same set of two reversible
oxidation waves; their respective half-wave potentials have a
value of 0.55 V/SCE and 0.74 V/SCE. They correspond to the
successive one-electron oxidation of the donor subunit leading
to the radical cation and the dication both centered in the
diaminofluorene part. These radical cations are not sensible to
the presence of oxygen since the same voltammograms are
obtained in aerated solution. Recently, it has been reported a
simple chemical method to obtain radical cations from
aromatic amines whose oxidation potentials are below
1.0 V/SCE.31,32 The reaction involves the mixing of aromatic
amine with copper salt in acetonitrile solution. Fig. 4 shows the
evolution of absorption spectrum of DAF in presence of
5 eq. of CuSO4 in acetonitrile, from the time-dependent change
observed in the absorption spectra (inset Fig. 4), two kinetic
steps are clearly identified. Consecutive to the rapid disappear-
ance of the absorption band of DAF, new bands are developing
in the 410–560 nm region and in the near infrared region
(l 4 900 nm), two isosbestic points are then displayed at
278 nm, 400 nm. This first step corresponds to the formation
of the radical cation. In the second step, the absorption band of
the radical is markedly decreasing with the concomitant
development of a new one in 600–900 nm range. This band is
ascribed to the absorption of the dication. The final solution is
stable for several hours indicating the persistent character of
both oxidized species. A strong stabilizing resonance effect is
assumed to explain such a persistent character. The similar
reactions are observed with DAFB.
Symmetry and electronic properties of excited states
Fig. 5 shows the normalized absorption spectra and the
normalized fluorescence spectra of DAFB in solvents of
increasing polarity. The Table 1 gathers the photophysical
data related to each emitting species. Despite a very slight blue
shift observed for the emission band of DAFB with respect to
the fluorescence of DAF, the two chromophores exhibit the
same fluorescence band shape. Moreover, in solvents with low
dielectric constant (e o 2.5), the fluorescence spectrum of
DAFB shows a very weak intensive band in the 500–650 range
(inset Fig. 5), the nature of this band will be further discussed.
ꢁc
This journal is the Owner Societies 2009
Phys. Chem. Chem. Phys., 2009, 11, 2622–2630 | 2625

View Article Online
Table 2 Calculated and observed electronic transitions of chromophoresa
DAF
DAFB
Eexp b/eV
Eth/eV
f
Pol.c
Eexp b/eV
Eth/eV
f
Pol.c
S0–S1
S0–S2
S0–S3
3.32
4.10
3.33
4.08
4.11
1.07
0.11
0.33
x
(y,z)
(y,z)
3.35
4.08
3.32
4.09
4.10
1.06
0.23
0.25
x
(y,z)
(y,z)
a
Eexp: Experimental value for the energy of the electronic transition; Eth: theoretical value for the energy of the electronic transition; Pol.: main
b
components for the polarization of the electronic transition. Values in acetonitrile. xyz axes are displayed in Scheme 1.
c
Fig. 2 Frontier molecular orbitals of DAF.
Fig. 4 Time-dependent changes in the absorption spectrum of DAF
(10ꢀ5 M) in presence of 5 eq. of CuSO4 (solvent = acetonitrile).
Inset: evolution of the absorbance at 480 nm and at 810 nm.
Fig.
3 Cyclic voltamogramms of DAF (full line) and DAFB
(dash-dotted line) in acetonitrile + (nBu)4NPF6 (0.1 M) on platinum
electrode at 200 mV sꢀ1 (concentration of chromophores: 10ꢀ3 M).
For both chromophores, the zero–zero transition (E00) lies at
3.18 eV and the emission band is hardly shifted to the red region
from hexane to DMSO (i.e. 980 cmꢀ1). The change in dipole
moment between the ground state and excited states, Dmg–e can
be estimated from the Dnmax difference between the maxima of
the absorption and fluorescence bands on the basis of Lippert’s
Df function and with the assumption that the Onsager cavity
has a radius of 5 A, value obtained from the fully geometry
optimization of DAF by DFT method. The change in dipole
moment from ground to excited state does not exceed 6 D, it
thus confirms the weak polar character of the emitting species.16
DAF and DAFB exhibit radiative rate constants (kr) of same
order of magnitude with mean values of ca. 0.39 ꢂ 0.05 ꢃ 109 sꢀ1
and 0.32 ꢂ 0.05 ꢃ 109 sꢀ1 respectively. It confirms the
electronic equivalency of each excited state. The radiative rate
constant value can be directly connected to the square of the
fluorescence transition moment Mf according to the following
equation:33,34
64p4
3h
kr ¼
n3n3f Mf2
Where n denotes the refractive index of the medium and nf the
maximum fluorescence wavenumber (cmꢀ1). The change of
the electronic or molecular geometry between the absorption
and fluorescence process can also be derived by comparison of
the radiative rate constant kr with Strickler–Berg rate constant
krSB calculated from equation:35
Z
8pc103 ln 10
NL
kSr B
¼
n2n3f eðnaÞdðln naÞ
ꢁc
This journal is the Owner Societies 2009
2626 | Phys. Chem. Chem. Phys., 2009, 11, 2622–2630

View Article Online
clearly indicate that the transition moments for the absorption
and for the emission are collinear with the same polarization
along the long axis of the 2,7-diaminofluorene moiety.
Moreover, DAFB and DAF show the same phosphorescence
spectrum with a maximum located at 515 nm which leads to a
triplet energy of ca. 2.41 eV (Fig. 6). In both cases, the
corresponding luminescence lifetime is 1.5 s which suggests
that the T1 states have a p,p* character. The average
phosphorescence anisotropy shows values roughly close to
0 in each case, this is consistent with an out-of-plane orienta-
tion of the emission moment usually observed for aromatic
systems.36,37 Indeed the theoretically spin-forbidden T1 - S0
transition is assumed to be strengthened through vibronic
mixing and singlet-state mixing with polarization normal to
the molecular plane.36
Fig. 5 Normalized absorption and fluorescence spectra of DAFB in
solvents of increasing polarity. Inset: fluorescence of the exciplex in
hexane.
Intramolecular quenching of singlet locally excited state
The electronic equivalency observed for the S1 states of DAF
and DAFB singularly contrasts with the large differences of
their respective fluorescence quantum yield Ffluo), for instance
In this previous relation, the integrated band absorption
spectrum is proportional to the square of absorption
transition moment (Ma) as follows:
F
fluo of DAF is 4 times higher as that of DAFB in hexane, and
becomes 10 times higher in acetonitrile. The comparison of the
non-radiative rate constants (knr) is also in line with these
observations: knr of DAFB is 5 times and 15 times higher than
that for DAF in hexane and in acetonitrile respectively. Hence,
the presence of the 3-methoxybenzophenone subunits in the
structure of DAFB should lead to additional non-radiative
deactivation pathways.
Z
8p3NLn
3hc103 ln 10
eðnaÞdðln naÞ ¼
Ma2;
where e(na) corresponds to the extinction coefficient at the
wavenumber na. Therefore the ratio between kr and krSB leads
to ratio of the squared fluorescence and absorption moments.
The calculation of krSB leads to a value of ca. 0.33 ꢃ 109 sꢀ1
and 0.34 ꢃ 109 sꢀ1 for DAF and DAFB, respectively. The
ratios Mf2/Ma2 exhibit values close to unity (i.e. 1.18 for DAF
and 0.94 for DAFB). Hence, the locally excited state (LE) of
each chromophore does not undergo any change in the
electronic or conformational structure. The presence of
crowded substituents at the rim of the diaminofluorene core
does not generate any sterical constraint that could affect the
electronic feature of the chromophore at the excited state.
Such an assumption is also confirmed by the measurement of
the fluorescence anisotropies whose spectra are displayed
in Fig. 6. DAF and DAFB exhibit an average value of
ca. 0.34 ꢂ 0.03 and 0.36 ꢂ 0.02, respectively. These results
Due to the high-lying proximity of the S1 state (3.18 eV)
with a p,p* character and the T1 state of the 3-methoxy
benzophenone (3.00 eV) which exhibits an n,p* character,29,30
a first hypothesis that could explain the quenching of singlet
state of DAFB should be a significant increase of the
intersystem crossing (ISC). Table 3 gathers FISC and kISC for
both molecules in apolar and polar solvent. We can notice that
kISC decreases by at least a factor two from hexane to
acetonitrile, this is clearly an opposite tendency as that
observed for knr. Even DAFB shows a kISC three times higher
than that of its model, the corresponding values are not high
enough to assert that ISC constitutes the main non-radiative
deactivation route, for instance kISC in acetonitrile only
represents 5% of knr.
The significant dependence of knr upon solvent polarity
suggests the occurrence of an intramolecular photoinduced
electron transfer mechanism (PIET) as a second hypothesis.
From the Rehm–Weller formalism, the free energy associated
to such a process can be estimated with the following equation:
DGeT = Eox ꢀ Ered ꢀ E00 + C
where Eox and Ered correspond to the oxidation and reduction
potential of the donor and acceptor respectively, E00 is the
Table 3
DAF
DAFB
FISC
kISC/10ꢀ9 sꢀ1
FISC
kISC/10ꢀ9 sꢀ1
Fig. 6 Superimposition of the emission spectrum and the emission
anisotropy of DAFB (lem = 370 nm) in glassy matrix of ethanol at
77 K.
HEX
ACN
0.21
0.14
0.20
0.08
0.17
0.05
0.59
0.26
ꢁc
This journal is the Owner Societies 2009
Phys. Chem. Chem. Phys., 2009, 11, 2622–2630 | 2627

View Article Online
energy of singlet excited state and C is the coulombic energy
term characterizing the interaction of the radical ion pairs.
This last term can be neglected in a highly polar solvent such
as acetonitrile. Assuming a PIET process from the 2,7-diamino-
fluorene to the 3-methoxybenzophenone moieties, the
estimated DGeT has a value of ca. ꢀ0.78 eV in acetonitrile.
Therefore PIET is thermodynamically allowed. This is also
consistent with the observed fluorescence quenching of DAF in
presence of increasing amount of benzophenone, the corres-
ponding Stern–Volmer coefficient has a value of ca. 52 Mꢀ1 in
acetonitrile. The quenching reaction leads to a concomitant
decrease of the fluorescence lifetime of DAF with a dynamic
quenching rate constant of ca. 3.3 1010 Mꢀ1 sꢀ1 which
indicates a diffusion-controlled process.
The occurrence of a PIET process is also supported by
the observation of the fully charge-separated state by
time-resolved absorption spectroscopy. Pump–probe measure-
ments were performed with excitation at 370 nm promoting
the diaminofluorene derivative to its first excited singlet state.
Fig. 7 displays the transient spectra of DAFB and its model in
acetonitrile at 5 ps and 900 ps delays after laser pulse. Both
chromophores initially exhibit the same transient absorption
bands with a maximum absorption wavelength located at
586 nm. The S1 - Sn absorption of DAF relaxed without
any spectral changes and the kinetics measured at 600 nm is
correctly fitted with a monoexponential (inset Fig. 7A). The
obtained decay time (i.e. 1.44 ꢂ 0.1 ns) is in close agreement
with the lifetime of the S1 state measured by time resolved
fluorescence. However, for DAFB, after the relaxation of the
S1 - Sn bands two weak bands are observed until the end
of kinetics. The first one which is centered at 505 nm is
Fig. 8 (A) Fluorescence spectrum of exciplex in toluene (circles)
obtained from subtraction of normalized emission spectra of DAFB
(full line) and DAF (dashed line). (B) Fluorescence decay behaviour of
DAFB in toluene at 580 nm (lexc = 372 nm). The Figure also shows
the residuals and the best monoexponential fit to the decay curve.
comparable to the band of the radical cation of DAF observed
in Fig. 4, the second band localized in 550–750 nm range can
be confidently attributed the absorption of the radical anion of
the 3-methoxy benzophenone.29,38,39 Two components can be
extracted from the kinetics measured at 650 nm (inset Fig. 7B),
a short-time component (i.e. 190 ps) ascribable to S1 state
relaxation and
a long-time one whose relaxation time
(i.e. 1.9 ns) has a value close the detection upper limit of the
instrument.
As previously mentioned, the fluorescence band of DAFB
shows a weakly intensive shoulder above 500 nm in low polar
solvents as hexane, di-n-butylether or toluene. Such a shoulder
is totally absent for DAF. A broad band is then revealed by
subtraction of the normalized emission spectrum of DAFB
with its analogue one as shown in Fig. 8A. This band is
solvent-dependent and its maximum is clearly red-shifted from
hexane to toluene (Dl = 20 nm). The excitation spectrum
measured at 580 nm matches the absorption spectrum of
DAFB indicating that this new emitting species arises from
the chromophore. Moreover, the relative intensity of the band
is invariant by increasing the concentration of DAFB from
2 ꢃ 10ꢀ7 M to 1 ꢃ 10ꢀ5 M which excludes any inter-molecular
effect. We can therefore ascribe this band to the emission of an
exciplex (E) stemming from the intramolecular charge transfer
between the diaminofluorene part and a 3-methoxy benzo-
phenone one. In toluene, the emission contribution of this
species correspond to 3% of the total emission which leads to a
very weak fluorescence quantum yield of ca. 5 ꢃ 10ꢀ4. As
shown in Fig. 8B, the fluorescence intensity collected at
580 nm in toluene decays monoexponentially with an emission
lifetime of 4.70 ns. Hence, a precursor–successor mechanism is
not evidenced since the decay does not exhibit a second
short lifetime component with a negative and opposite pre-
exponential factor.33 This suggests that the emitting exciplex is
not directly provided from the relaxed LE state. Moreover the
monoexponential time decay observed for the exciplex
indicates that the emission arises from one type of conformers,
Fig. 7 Transient absorption spectra of DAF (top) and DAFB
(bottom) in acetonitrile. Insets: transient absorption kinetics at
600 nm (top) and at 650 nm (bottom).
ꢁc
This journal is the Owner Societies 2009
2628 | Phys. Chem. Chem. Phys., 2009, 11, 2622–2630

View Article Online
generally a parallel sandwich configuration is promoted since
it maximizes the overlap of the two charge clouds. From the
full geometrical optimized structure of DAFB which is
depicted in Fig. 9, we can observe a sandwich-like conforma-
tion structure between the diaminofluorene moiety and a
benzophenone group with a closest inter-aromatic distance
of ca. 6.4 A. The combining factors of large twist angles
between aromatic rings linked to the nitrogen atom and the
connection on the meta position to benzophenone moiety
should counterbalance the unfavourable role of the short alkyl
linkage. Hence, this structural pre-organization at the ground
state should prevent a large amplitude geometrical relaxation
at excited state and promote the occurrence of the exciplex
emission in low polar solvent. The fluorescence of this exciplex
is largely quenched in more polar solvent. This can be ascribed
to the formation of a non-fluorescent geminate or solvent-
shared ion-pairs.40
Fig. 10 Temperature effect on the fluorescence spectra of DAF and
DAFB in toluene. Inset: invariance of the fluorescence of exciplex
upon cooling.
Fig. 10 shows the temperature effect on the fluorescence
spectra of each compound. The fluorescence of DAFB is
clearly more temperature-dependent than its model
fluorophore. Decreasing temperature from 100 1C to 0 1C
leads to a fluorescence enhancement of DAFB by a factor of
3.3 while this factor hardly reaches a value of 1.2 for DAF.
Interestingly, the fluorescence enhancing of DAFB contrasts
with the invariance of the exciplex emission (inset to Fig. 10), it
confirms the absence of a parent–daughter relationship
between the LE and E species. The solvent shell reorganization
consecutive to PIET leads the radical–ion pair stabilization.
Hence the slow down of solvation dynamics and of the
geometrical relaxation at low temperature should hamper
PIET leading to the ‘switch on’ of emission DAFB. Assuming
that PIET constitutes the major deactivation pathway, we can
derive the relationship:
keT
kr0
1
1
Fig. 11 Arrhenius plot for the substation of the reciprocal quantum
yields of DAFB and DAF (solvent = toluene).
¼
ꢀ
F
F0
Where keT denotes the rate constant PIET, kr0 is the radiative
constant of DAF. F0 and F are fluorescence quantum yields of
DAF and DAFB, respectively. Only keT is temperature
dependent and with the assumption of an Arrhenius relation-
Ea
ship (i.e. keT ¼ A expðꢀ Þ), activation energy can be
RT
measured by temperature effect. Fig. 11 shows the linear
1
1
1
relationship between lnðF ꢀ F Þ versus T. The activation energy
0
related to the association reaction (Ea) is obtained from the
slope of the straight line. Hence in toluene, activation energy
of ca. 9.0 kJ molꢀ1 is obtained. The value clearly corresponds
to the ‘‘activation energy’’ (EZ) of the viscosity of toluene41
(i.e. 8.3 kJ molꢀ1) and suggests that the rate constant of PIET
varies with the temperature in the same way as solvent
viscosity. Therefore PIET does not proceed over a significant
activation barrier and does not require a large-amplitude
conformational change from the locally excited state.
Conclusion
We have characterized the photophysical properties of a
2,7-bis-(N-4-methoxyphenyl-N-phenylamino)fluorene deriva-
tive covalently linked to two 3-methoxy benzophenone
Fig. 9 Optimized ground-state structure of DAFB obtained by DFT
calculation.
ꢁc
This journal is the Owner Societies 2009
Phys. Chem. Chem. Phys., 2009, 11, 2622–2630 | 2629

View Article Online
subunits with short flexible spacers. At ground state,
the comparison with a model system clearly indicates that
the presence of the benzophenone subunits at the rim of the
donor core does not affect any of the spectroscopic or electro-
nic characteristics. However at the excited state, a photo-
induced intramolecular electron transfer leads to an efficient
quenching of the locally excited state. In low polar solvent,
the emission of an exciplex confirms the occurrence of a charge
transfer process which is also corroborated by the fluorescence
switch ‘on’ of the chromophore at low temperature.
The temperature effect also indicates that this quenching
pathway does not cross a potential barrier. Hence the
conformational preorganization at ground state leads to a
weak geometrical relaxation at excited one.
18 B. A. Reinhardt, L. L. Brott, S. J. Clarson, A. G. Dillard,
J. C. Bhatt, R. Kannan, L. Yuan, G. S. He and P. N. Prasad,
Chem. Mater., 1998, 10, 1863–1874.
19 C. Martineau, R. Anemian, C. Andraud, I. Wang, M. Bouriau and
´
P. L. Baldeck, Chem. Phys. Lett., 2002, 362, 291–295.
20 K. D. Belfield, K. J. Schafer, W. Mourad and B. A. Reinhardt,
J. Org. Chem., 2000, 65, 4475–4481.
21 J. F. Hartwig, M. Kawatsura, S. I. Hauck, K. H. Shaughnessy and
L. M. Alcazar-Roman, J. Org. Chem., 1999, 64, 5575–5580.
22 R. Meech and D. Phillips, J. Photochem., 1983, 23, 193–217.
23 D. V. Connor and D. Phillips, Time Correlated Single Photon
Counting, Academic Press, London, 1984.
24 T. J. V. Prazeres, A. Fedorov, S. P. Barbosa, J. M. G. Martinho
and M. r. N. Berberan-Santos, J. Phys. Chem. A, 2008, 112,
5034–5039.
25 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski,
J. A. Montgomery, Jr., R. E. Stratmann, J. C. Burant,
S. Dapprich, J. M. Millam, A. D. Daniels, K. N. Kudin,
M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi,
R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford,
J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui,
K. Morokuma, P. Salvador, J. J. Dannenberg, D. K. Malick,
A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. Cioslowski,
J. V. Ortiz, A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox,
T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara,
M. Challacombe, P. M. W. Gill, B. G. Johnson, W. Chen,
M. W. Wong, J. L. Andres, C. Gonzalez, M. Head-Gordon,
E. S. Replogle and J. A. Pople, GAUSSIAN 98 (Revision A.11),
Gaussian, Inc., Pittsburgh, PA, 2001.
References
1 A. Kapturkiewicz and J. Nowacki, J. Phys. Chem. A, 1999, 103,
8145–8155.
2 A. Kapturkiewicz, J. Herbich, J. Karpiuk and J. Nowacki, J. Phys.
Chem. A, 1997, 101, 2332–2344.
3 M. R. Wasielewski, Chem. Rev., 1992, 92, 435–461.
4 T. Hirsch, H. Port, H. C. Wolf, B. Miehlich and F. Effenberger,
J. Phys. Chem. B, 1997, 101, 4525–4535.
5 M. Niemi, N. V. Tkachenko, A. Efimov, H. Lehtivuori,
K. Ohkubo, S. Fukuzumi and H. Lemmetyinen, J. Phys. Chem.
A, 2008, 112, 6884–6892.
6 H. Imahori, N. V. Tkachenko, V. Vehmanen, K. Tamaki,
H. Lemmetyinen, Y. Sakata and S. Fukuzumi, J. Phys. Chem. A,
2001, 105, 1750–1756.
7 J. W. Park, B. A. Lee and S. Y. Lee, J. Phys. Chem. B, 1998, 102,
820–8215.
26 K. D. Belfield, M. V. Bondar, J. M. Hales, A. R. Morales,
O. V. Przhonska and K. J. Schafer, J. Fluoresc., 2005, 15, 3–11.
27 P. Najechalski, Y. Morel, O. Ste
Phys. Lett., 2001, 343, 44–48.
28 J.-F. Briere and M. Cote
3123–3129.
´
phan and P. L. Baldeck, Chem.
´
, J. Phys. Chem. B, 2004, 108,
8 Z. R. Grabowski, K. Rotkiewicz and W. Rettig, Chem. Rev., 2003,
103, 3899–4032.
29 A. C. Bhasikuttan, A. K. Singh, D. K. Palit, A. V. Sapre and
J. P. Mittal, J. Phys. Chem. A, 1998, 102, 3470–3480.
9 T. P. Le, J. E. Rogers and L. A. Kelly, J. Phys. Chem. A, 2000, 104,
6778–6785.
30 F. Bosca, G. Cosa, M. A. Miranda and J. C. Scaiano, Photochem.
´
Photobiol. Sci., 2002, 9, 704–708.
10 D. Gust, T. A. Moore and A. L. Moore, Acc. Chem. Res., 1993, 26,
198–205.
31 M. Kirchgessner, K. Sreenath and K. R. Gopidas, J. Org. Chem.,
2006, 71, 9849–9852.
11 M. Isosomppi, N. V. Tkachenko, A. E, H. Vahasalo, J. Jukola,
P. Vainiotalo and H. Lemmetyinen, Chem. Phys. Lett., 2006, 430,
36–40.
12 J. M. Warman, K. J. Smit, M. P. D. Haas, S. A. Jonker,
M. N. Paddon-Row, A. M. Oliver, J. Kroon, H. Oevering and
J. W. Verhoeven, J. Phys. Chem., 1991, 95, 1979–1987.
13 M. N. Paddon-Row, A. M. Oliver, J. M. Warman, K. J. Smit, M.
P. D. Haas, H. Oevering and J. W. Verhoeven, J. Phys. Chem.,
1988, 92, 6958–6962.
14 G. L. Closs and J. R. Miller, Science, 1988, 240, 440–447.
15 J. M. Hales, D. J. Hagan, E. W. V. Stryland, K. J. Schafer,
A. R. Morales, K. D. Belfield, P. Pacher, O. Kwon, E. Zojer and
J. L. Bredas, J. Chem. Phys., 2004, 121, 3152–3160.
16 K. D. Belfield, M. V. Bondar, O. V. Przhonska, K. J. Schafer and
W. Mourad, J. Lumin., 2002, 97, 141–146.
32 S. Sumalekshmy and K. R. Gopidas, Chem. Phys. Lett., 2005, 413,
294–299.
33 J. B. Birks, Photophysics of Aromatic Molecules, Wiley-
Interscience, New York, 1970.
34 M. Maus, W. Rettig, D. Bonafoux and R. Lapouyade, J. Phys.
Chem. A, 1999, 103, 3388–3401.
35 S. J. Strickler and R. A. Berg, J. Chem. Phys., 1962, 37, 814.
36 J. Catalan, Chem. Phys., 2005, 316, 253–259.
´
37 M. P. Lettinga, H. Zuilhof and M. A. M. J. v. Zandvoort, Phys.
Chem. Chem. Phys., 2000, 2, 3697–3707.
38 C. Devadoss and R. W. Fessenden, J. Phys. Chem., 1990, 94,
4540–4549.
39 T. Shida, S. Iwata and M. Imamura, J. Phys. Chem., 1974, 78,
741–748.
40 N. Ghoneim, Spectrochim. Acta, Part A, 2001, 57, 483–489.
41 M. J. Assael, N. K. Dalaouti and J. H. Dymond, Int. J.
Thermophys., 2000, 21, 291–299.
17 K. D. Belfield, M. V. Bondar, F. E. Hernandez, O. V. Przhonska
and S. Yao, J. Phys. Chem. B, 2007, 111, 12723–12729.
ꢁc
This journal is the Owner Societies 2009
2630 | Phys. Chem. Chem. Phys., 2009, 11, 2622–2630